Neuroblastoma, a tumor of neural crest origin, is the most common extra-cranial solid tumor of childhood. Amplification of MYCN, a MYC family ontogeny, occurs in ~25% of neuroblastoma and marks high-risk disease. Neuroblastoma is unique among solid tumors in that all patients typically respond to initial chemotherapy. Risk is thus associated with relapse, which is typically therapy-resistant. Contributing to this resistance a large proportion of relapsed tumors have therapy-induced mutations in p53 pathway genes, not evident at initial presentation. Mirroring human disease, the TH-MYCN mouse model of neuroblastoma is responsive to chemotherapy. In order to provide a mouse model of chemotherapy resistant, relapsed disease, we have crossed TH-MYCN mice to mice knocked-in for p53ER, which are p53 deficient at baseline. Exposure to hydroxytamoxifen allows rapid and reversible p53 restoration. Resulting neuroblastoma tumors show increased penetrance and decreased latency at baseline. When p53 is restored, mice show initially improved survival followed by relapse. In Aim 1, I will characterize how tumors evade reactivation of p53 to enable. I will then characterize the importance of p53 in chemotherapy resistance to existent conventional and targeted therapies, establishing a baseline for pre-clinical development of novel therapeutic agents, including those generated in Aim 2. The stability of MYCN protein in neuroblastoma is tightly regulated by upstream signaling through the PI3K/mTOR pathway as well as by a kinase-independent scaffolding function of the Aurora A protein. In Aim 2, I will characterize a novel class of Type II inhibitors of Aurora A kinase designed specifically to alter the secondary structure of Aurora A and disrupt the kinase-independent stabilization of MYCN. In pilot studies I have identified candidate inhibitors that dramatically decrease MYCN protein in neuroblastoma cells and that appear to do so by the destabilization mechanism proposed. I will complete structure activity relationship studies to further refine these candidate compounds. Successful completion will result in both an improved model of chemo-resistant neuroblastoma as well as a new class of targeted Aurora A inhibitors, predicted to be both highly potent and specific against neuroblastoma.